Aromatic pentafluoro-λ6-sulfanyl (SF5) surfactants: m-SF5(CF2)nC6H4SO3K

Mendeleev Communications Mendeleev Commun., 2006, 16(3), 182–184

Aromatic pentafluoro-O6-sulfanyl (SF5) surfactants: m-SF5(CF2)nC6H4SO3K Scott W. Winner,a Rolf W. Winter,a Jeremy A. Smith,a Gary L. Gard,*a Nicole A. Hannah,a Shankar B. Rananavare,a Barbora Piknovab and Stephen B. Hallb a b

Department of Chemistry, Portland State University, Portland, Oregon 97207, USA. Fax: + 1 503 725 9525; e-mail: [email protected] Department of Medicine, Oregon Health & Science University, Mail code NRC-3, Portland, Oregon 97239, USA DOI: 10.1070/MC2006v016n03ABEH002315

The synthesis and characterization of new fluorinated surfactants based on SF5-aromatic system are presented for the first time. For over 40 years,1 the only known SF5-aromatic systems contained the SF5C6H4 moiety. The new synthetic strategy that we have recently developed2 has opened the floodgate for preparing a large assortment of o, m, p-SF5(CF2)nC6H4X and chiral o, m, p-SF5CF2CFYC6H4X derivatives that have technologically important applications in lithium batteries, fuel cells and ionexchange resins.3–8 More readily degradable than CF3-terminated molecules, the SF5-aromatic structural motif is readily integrated into fluorinated polymers,2 liquid crystals9 and surfactants. In this paper, we focus on sulfonate surfactants, showing below (Scheme 1) how other classic analogues of cationic (CTAB: hexadecyltrimethylammonium bromide), anionic (SDS, sodium dodecyl sulfonate), and nonionics (Tritons: t-octylphenoxypolyoxyethanols) may be synthesised. A number of surface active SF5-fluoroalkyl sulfonic acids and salts are known.3,4,7,8,10,11 3M researchers first demonstrated that the polar, yet weakly hydrophobic SF5-terminated alkyl group exhibits surface activity greater than CF3-terminated compounds.10,11 They observed a lower surface tension for the 182

Mendeleev Commun. 2006

cationic (e.g. CTAB) H2CO Y = NH 2 MeBr (C H O)

2 4 m nonionic SF5(CF2)n (e.g. Tritons) Y = OH

Y

ClSO3H Y=H

anionic (e.g. SDS)

n = 2–10

Scheme 1 Synthetic approach.

shorter chain SF5-ammonium carboxylate than the longer chain CF3(CF2)6C(O)ONH4.10,11 While little is known concerning the surface tension of aqueous solutions of organic SF5-carboxylate salts; to our knowledge, nothing is known about organic SF5sulfonate salts. Since some of the lowest air–water surface tensions (15–20 mN m–1) are reported for para-fluorocarbon aryl sulfonates,12–15 we explored a new homologous series containing CF2 groups as a spacer between the SF5 group and

Mendeleev Commun., 2006, 16(3), 182–184 Table 1 Summary of thermodynamic parameters. C10(H) and C8(F) refer to C10H21C6H4SO3Na and C8F17SO3Li, respectively. The numbers in parentheses for a and b indicate the uncertainty. The numbers in parentheses for CMC and 'G correspond to values evaluated by NMR. (CF2)n

a

b

CMC/ 10–3 mol dm–3

G/10–3 mol per 1000 m2

s/Å2 molec–1

CMC/C20

PCMC/ mN m–1

–'Gmicelle/ kJ mol–1

–'Gads/ kJ mol–1

2 4 6 C10(H) C8(F)

20(1) 7(1) 0.77(1) — —

0.50(3) 0.45(5) 0.45(1) — —

37.6 (58.2) 18.1 (10.7) 1.9 3.1 6.3

3.0 3.5 4.1 3.9 3.0

55.8 46.4 41.0 43.0 55.0

2.1 2.7 2.5 1.4 10

38.2 41.0 40.3 25.8 42

17.8 (17.0) 19.6 (21.2) 25.2 26.3 22.3

29.7 (30.7) 31.1 (32.4) 35.2 33.0 36.2

aromatic sulfonic acid/salt. Remarkably, we find evidence of aggregation and interfacial adsorption (hence surface tension reduction) even in meta-SF5(CF2)nC6H4SO3K (n = 2, 4, 6). The acids and salts m-SF5(CF2)nC6H4SO3H and m-SF5(CF2)nC6H4SO3K (n = 2, 4, 6), were synthesised (see Scheme 1) by sulfonating SF5(CF2)nPh with chlorosulfonic acid followed by a base hydrolysis to yield the sulfonate salts. Removing K+ ions by ion-exchange chromatography gave the corresponding acid (see Supplementary Materials for synthetic and structural details, which are available free via http://www.turpion.org/suppl/mc/ 2315/suppl2315.pdf). Surface tension, g, of these compounds [Figure 1(a)] decreases with sub-phase surfactant concentration, c, exhibiting a distinct leveling after a certain critical concentration that depends on the fluorocarbon chain length. The minimum surface tension is approximately 32 mN m–1, higher than the para-fluorocarbon aryl sulfonates. The concentration dependence of surface tension is consistent with an interfacial adsorption punctuated either by micellization or by exceeding the solubility limit, i.e., when the Krafft temperature is > 23 °C. To distinguish between the two possibilities, NMR chemical shift studies were conducted. Micellization leads to a continuous variation of chemical shift with concentration.16 Exceeding the solubility limit, however, 80

(a) n=2 n=4

60 g /mN m–1

n=6

40

20 –11

–8

–5

–2

ln c

–16

46.2

∆Gmicelle /kJ mol–1

(b) n=4

45.6 d /ppm

results in a two-component NMR spectrum arising from solidliquid coexistence due to slow (on the time scale of NMR) molecular exchange between the phases. The chemical shifts for axial SF5 and CF2 groups shown in Figure 1(b) display a smooth variation with surfactant concentration. From the bend in the slopes of surface tension and chemical shift data, we estimated the critical micelle concentrations (CMCs). The surface pressure at the CMC, PCMC (= gw – gCMC),17 measures the surfactant effectiveness in reducing surface tension. Present surfactants display significantly higher (lower) PCMC than the protonated alkyl-aryl sulfonates (p-fluoroaryl sulfonates). The ratios of adsorption to micellization efficiencies, CMC/C20, (where C20 is a surfactant concentration needed to reduce gw by 20 mN m–1) are greater (smaller) than the protonated (fluorinated) surfactants. This may be due to inefficient interfacial packing resulting from the meta-placement of the hydrophilic sulfonate group off the long molecular axis, rectifiable in para-derivatives. To gain further insight into the self-assembly process, interfacial packing parameters were determined using Langmuir– Szyszkowski (LS) and Gibbs models.18 The LS model predicts that g = g0[1 – bln(1 + c/a)], where g0 is the surface tension of water. The coefficient b = RTG¥/g0, where G¥ is the maximum possible adsorption, is a characteristic constant of a homologous series while a is compound-specific. Nonlinear least squares fits to the g vs. c data (Table 1) show a constant b (G¥ » 7×10–3 mol/1000 m2) and a systematic decrease in a with number of CF2 groups. The Gibbs interfacial excess,19 G = (0.5RT)dg/dln c,† and molecular area s (=1023/(NAG in units of nm2) were also computed using the semi-log plot of g vs. ln c [Figure 1(a)]. Note that G/G¥ (» 0.5) verifies loose interfacial packing, which depends on the fluorocarbon chain length. Longer chains, perhaps due to favourable fluorocarbon– fluorocarbon interactions, pack more closely. The micellization and adsorption free energies, 'Gmic » RTln (cCMC /55.5) and 'Gads = 'Gmicelle – PCMC s, respectively, are presented in Table 1. These free energies, cross-sectional areas and surface excesses are comparable to C7F15SO3Li and equivalent (using 1.5CH2º1CF2 rule19) protonated p-C10H21C6H4SO3Na surfactants. Figure 2 indicates that the decrease in 'Gmicelle

45.0

–21

–26

44.4

1

3

5

7

Number of CF2 groups 43.8 0.4

0.9

1.4

1.9

log c Figure 1 Semi-log plots of (a) surface tension and (b) fluorine chemical shift vs. surfactant concentration for varying numbers of CF2 groups. In (b) from bottom to top, we present (CF2)n and equatorial (doublet) fluorine from the SF5 group, respectively. The chemical shift of the CF2 group has been shifted by an additive constant (154.7 ppm) and the chemical shift of the equatorial SF5 , by 1.4 ppm for clarity of presentation.

Figure 2 Plot of 'Gmicelle vs. the number of CF2 groups attached. The fitted line gives the slope and intercept of –1.9 kJ per CF2 group and –13 kJ mol–1, respectively. †

The factor of 2 accounts for a complete ionization of the potassium salt. When this ionization is not considered, the extracted values of G > G¥ are physically impossible. Resonance structures of SF5 and CF2 groups attached to phenyl rings predict an electron deficiency at o- and p-positions, stabilising the ionised sulfonate group.20

Mendeleev Commun. 2006

183

Mendeleev Commun., 2006, 16(3), 182–184

per CF2 group is considerably smaller (~2 kJ mol–1) than for typical fluorinated surfactants,19 but comparable to values for CH2 groups.21 Parallel free energy reduction is observed in RCOO(CH2)nSO3Na and mixed fluorohydrocarbon nonionics where the presence of two highly polar opposing groups reduces the 'Gmicelle(CH2) by half.18 Structurally similar bolaamphiphiles22 containing two terminal polar hydrophilic groups form vesicles rather than micelles. The polar hydrophobic SF5 terminus, however, appears to suppress vesicular structures, but it may permit microemulsions of polar oils such as asphaltenes, amphiphatic proteins and lyotropic liquid crystals, which are valuable templates for the synthesis of nanostructured material. This work was supported by the National Science Foundation (CHE-9904316), the Petroleum Research Foundation (ACS-PRF no. 34624-AC7) and the National Institute of Health (HL54209). We are grateful to J. Morré (Mass Spectrometry Laboratory, Oregon State University, Corvallis, Oregon) for obtaining the high-resolution mass spectra (HRMS). References 1 (a) G. H. Cady, in Some Recent Advances in Fluorocarbon Chemistry, Proceedings of the Chemical Society, April 1960, pp. 133–138; (b) R. K. Pearson and R. D. Dresdner, in A Preparation of Primary Perfluoroalkylamines, 140th Meeting of the American Chemical Society, Chicago, 1961, pp. 4743–4746; (c) W. A. J. Sheppard, J. Am. Chem. Soc., 1962, 84, 3064; (d) R. W. Winter, R. A. Dodean and G. L. Gard, in FluorineContaining Synthons, ed. V. A. Soloshonok, ACS Symposium Series 911, American Chemical Society, Washington, DC, 2005, pp. 87–119. 2 R. W. Winter, R. A. Dodean, J. A. Smith, R. Anilkumar, D. J. Burton and G. L. Gard, J. Fluorine Chem., 2005, 126, 1202. 3 J. M. Canich, M. M. Ludvig, G. L. Gard and J. M. Shreeve, Inorg. Chem., 1984, 23, 4403. 4 G. L. Gard, A. Waterfeld, R. Mews, J. Mohtasham and R. W. Winter, Inorg. Chem., 1990, 29, 4588.

184

Mendeleev Commun. 2006

5 N. N. Hamel, P. G. Nixon, G. L. Gard, R. L. Nafshun and M. M. Lerner, J. Fluorine Chem., 1996, 79, 81. 6 R. L. Nafshun, M. M. Lerner, N. N. Hamel, P. G. Nixon and G. L. Gard, J. Electrochem. Soc., 1995, 142, L153. 7 R. J. Terjeson, J. Mohtasham and G. L. Gard, Inorg. Chem., 1988, 27, 2916. 8 R. J. Willenbring, J. Mohtasham, R. W. Winter and G. L. Gard, Can. J. Chem., 1989, 67, 2037. 9 J. A. Smith, R. A. DiStasio, Jr., N. A. Hannah, R. W. Winter, T. J. R. Weakley, G. L. Gard and S. B. Rananavare, J. Phys. Chem. B, 2004, 108, 19940. 10 J. Hansen and P. M. Savu, U.S. Patent 5,159,105, 1992. 11 J. Hansen and P. M. Savu, U.S. Patent 5,286,352, 1994. 12 A. M. Hodges, R. W. Winter, S. W. Winner, D. A. Preston and G. L. Gard, J. Fluorine Chem., 2002, 114, 3. 13 J. Hutchinson, Fette Seifen Anstr. V., 1974, 76, 158. 14 N. Ishikwaw and M. Sasbe, J. Fluorine Chem., 1984, 25, 241. 15 D. Prescher, B. Dreher, W. Dmowski and R. Wozniacki, German Patent (East) 239,789, 1986. 16 B. Lindman, G. Lindblom, H. Wennerstoem and H. Gustavsson, in Ionic Interactions in Amphiphilic Systems Studied by NMR, Plenum, New York, 1977, vol. 1, p. 195. 17 M. J. Rosen, Surfactants and Interfacial Phenomena, 2nd edn., Wiley, New York, 1989. 18 A. Adamson and A. Gust, Physical Chemistry of Surfaces, 6th edn., Wiley, New York, 1997, vol. 1, pp. 67. 19 V. M. Sadtler, F. Giulieri, M. P. Krafft and J. G. Riess, Chem. Eur. J., 1998, 4, 1952. 20 W. A. Sheppard, J. Am. Chem. Soc., 1962, 84, 3072. 21 C. Tanford, in The Hydrophobic Effect, 2nd edn., Wiley, New York, 1980, pp. 42–59. 22 C. Di Meglio, S. B. Rananavare, S. Svenson and D. H. Thompson, Langmuir, 2000, 16, 128.

Received: 30th January 2006; Com. 06/2657